Acceptability and utilization of a lipid-based nutrient supplement formulated for pregnant women in rural Niger: a multi-methods study

Parent trial

In 2012, a phase III randomized, double-blind, placebo-controlled trial to test the efficacy of the bovine rotavirus-pentavalent vaccine (BRV-PV) began in Madarounfa, Niger (ClinicalTrials.gov Identifier: NCT02145000). Details of the parent trial and study setting have been published elsewhere [13]. In recognition of the need to identify potential boosters of immunogenicity in countries where oral vaccine efficacy is low, a cluster randomized sub-study was designed to identify complementary interventions that may increase vaccine efficacy. One potential mechanism hypothesized to underlie lower immunogenicity observed in some settings is related to the inhibitory effects of poor nutrition on immune development [14]. The role of nutrition in immune development has been reasonably well documented in observational studies, where both macro- and micronutrient deficiencies have been shown to result in the impairment of various components of the immune system [15]. There is also evidence that micronutrient, fatty acid and protein-energy nutrition in early life may play a role in immune development [16, 17], but there have been few randomized studies to evaluate the impact of early nutritional intervention on infant immune development. We therefore proposed to evaluate the effect of prenatal nutrition supplementation on infant immune response as a nutrition-immunogenicity sub-study; the sub-study specifically aimed to test whether daily prenatal nutritional supplementation with LNS or MMN improves infant immune response to three doses of oral vaccine, compared to IFA.

LNS formulation

Recruitment and inclusion criteria

This study was conducted in Madarounfa, a rural department in the Maradi region of Niger. The department comprises several communes, one of which comprises the urban area of Madarounfa. Recruitment took place in June (Ramadan) and again in September (non-Ramadan) 2014 among women presenting for routine antenatal care visits at one study health center. Women at least 18 years of age with a confirmed pregnancy were eligible to participate. Women were excluded if they had a known intolerance to milk or peanuts, or their clinical status at the time of evaluation required inpatient referral. Women who met the inclusion criteria and did not fit the exclusion criteria were given oral and written details of the study and asked to sign or thumbprint the printed informed consent form. After providing consent, participating women were asked to return to the health center the following day. Gestational age of participants was not assessed at the time of enrollment.

Study design and procedures

This study collected data during two calendar periods: June (Ramadan) and September (non-Ramadan) 2014. Ramadan is the Muslim holy month marked by daily fasting from sunrise to sunset. Pregnant women may be exempt from daily fasting, however household practices may influence individual food preparation and intake during this period. In Niger, where more than 98% of population is estimated to practice Islam, this design allowed for a description of Epi-E acceptability and utilization during two important periods when dietary intake and practices are very different for the majority of the population. In both time periods, a two-part, multi-methods study design was used among two independent samples of participants.

Part I was designed to evaluate Epi-E acceptability through repeated observation of two test meals provided at the health center. Prior to each test meal, trained study staff collected morbidity information from each participant using a self-report of the previous 24 h. Study staff combined Epi-E and locally-purchased ingredients into a fermented maize porridge mixture in a 1:4 ratio. Maize porridge was used in the test meal as it is a common, staple-food preparation available to and often consumed by pregnant women in this context. Each study participant was individually given 50 g of the Epi-E/maize porridge mixture while being observed by a trained interviewer outside of the health center. This amount represented 10 g of Epi-E (approximately two teaspoons) and 40 g of maize porridge (approximately 2–3 tablespoons).

Interviewers weighed the amount of Epi-E/maize porridge mixture that each participant was served and the amount remaining after 15 min. The portion consumed was defined as the difference between the starting and ending volume to the nearest gram. After the test meal, the interviewer asked each participant to rate the characteristics of the Epi-E/maize porridge mixture based on her perceptions of its organoleptic properties, including color, odor, taste, and overall liking, using a hedonic scale with 5 options (1 = Dislike a lot, 2 = Dislike a little; 3 = Neither like nor dislike; 4 = Like a little; 5 = Like a lot). The 5-point scale was visually depicted using a range of emoticon faces (very unhappy to very happy) that have been previously used to measure food acceptability in illiterate populations [20–22]. Participants were also asked open-ended questions for further explanations and clarifications. The test meal was repeated on two consecutive days to allow women to habituate to the procedures and as a strategy to reduce influence of observation on behaviors.

Part II followed Part I and was designed to simulate more realistic conditions outside of the health center by including a home-feeding trial where participants received a take-home ration of Epi-E to use. A one-day supply of Epi-E included one 40 g sachet to be divided into two meals. Each participant was given a 14-day supply; in the first home trial conducted in June, participants were provided two additional sachets (16 total) in consideration of potential sharing (e.g. distribution of Epi-E among individuals other than the intended beneficiary, including children and other adults within and outside of the household). Participants received oral instructions regarding the appropriate daily dose of Epi-E for consumption, proper storage at room temperature, and the importance of adding it to a meal only after the cooking process.

After completion of the 14-day study period, participants were invited to return to the health center. To estimate compliance, study staff recorded the number of unused sachets returned to the health center with each participant and asked each woman how many sachets were either consumed by the participant or shared with other individuals. A post-trial questionnaire, with both close- and open-ended questions was administered to further assess the incidence of adverse events (e.g. diarrhea, vomiting, fever, reduced appetite, and constipation), as well as the quantity, timing and mode of consumption of Epi-E during the 14-day home trial. In September, interviewers additionally asked participants about their perceptions of Epi-E properties using a hedonic scale, as well as their willingness to use Epi-E again and pay for it in the future after the home trial period. All interviewers provided textual field notes regarding their perception of the product acceptability for further triangulation of data from the interviewers’ perspective.

A sample size of 25 women provided more than 90% power to test the hypothesis that mean consumption in Part 1 was at least 50% of the Epi-E/maize porridge mixture offered, assuming a standard deviation (SD) of consumption of 30% of the amount offered and a true mean of 75%.

Data analysis

Participant characteristics, including socio-demographic characteristics and self-reported morbidity, as well as results from the Part 1 test meals and Part 2 home trial were summarized using means (SD) for continuous measures, and using counts and proportions for discrete measures. A one-sample t-test was used to assess the difference between the hypothesized and observed mean consumption [23]. Hedonic scale responses rating sensory properties were presented as medians (min, max). Textual responses generated from the open-ended questions underwent an inductive review to identify thematic areas [24]. We considered both their frequency of mention and importance relative to the study aims when deeming emergent themes to be salient. Exemplar quotations were presented in text to illustrate key findings and as corroboration of quantitative results [25, 26]. All quantitative data were double-entered into an EpiData 3.1 database (EpiData, Odense, Denmark) and analyzed using Microsoft 2010 Excel (Microsoft, Redmond, Washington).

This study was conducted according to the guidelines laid down in the Declaration of Helsinki and all procedures involving human subjects/patients were approved by the National Consultative Ethics Committee of Niger, as well as the Western Institutional Review Board. Written informed consent was obtained from all participants.

Test meal results

14-day home trial results

In June (Ramadan), 72% of women ate half of one sachet at each serving, with servings most often taken in the morning and again later in the evening. By contrast, in September (non-Ramadan), women more often ate one full sachet each day, in either the morning (60.0%) or midday (40.0%), but never in the evening. Follow-up questions revealed that the preferred time of consumption was dependent on individual-level factors. Participants in both periods indicated that they were “hungrier” or had “big appetites” in the morning. Others chose when to eat Epi-E based on practical reasons related to work schedules and convenience. One person explained that she specifically chose to eat Epi-E at specific times, “to keep children from seeing [her eating it].” During June (Ramadan), participants reported consuming Epi-E early in the morning and/or around dinner or evening time, invoking dietary practices during the holy month for these preferred times, “Because of the fast, I prefer to take it [Epi-E] at dawn and after dinner.” In both time periods, Epi-E was almost universally consumed directly from the sachet. The most commonly reported reason for doing so was based on flavor. Many women said that they “like the taste alone, without the mixture”. Another salient theme was the notion that eating out of the sachet represented “direct” consumption, which was explained by participants to be both “good” and “better” than mixing. Women also cited convenience for using this mode of consumption.

Willingness to pay data were collected from participants in September (non-Ramadan). Nearly all women (95.7%) said they would purchase Epi-E if sold on the market. Women estimated that they would be willing to pay on average (SD) $0.31 (0.29) per sachet, compared to $0.05 (0.05), the average amount last paid for a daily prenatal micronutrient supplement and $0.31 (0.19), the average amount last paid for a serving of their favorite snack food.